Data analysis: The null ritual and the use of effect sizes

Conzelmann, Achim; Raab, Markus (2009). Data analysis: The null ritual and the use of effect sizes. Zeitschrift für Sportpsychologie, 16(2), pp. 43-54. Göttingen: Hogrefe 10.1026/1612-5010.16.2.43

[img] Text
fulltext.pdf - Published Version
Restricted to registered users only
Available under License Publisher holds Copyright.

Download (367kB) | Request a copy

This article analyzes the use of statistical tests in German-language sport psychology. Results from comparable analyses in, for example psychology, reveal discrepancies between theoretical requirements and empirical practice that have not yet been described or discussed in the field of sport psychology. Volumes 1994–2007 of the journal Zeitschrift für Sportpsychologie (formerly psychologie und sport) were analyzed to see whether research questions were formulated, how participants were sampled, what test theory was used, what significance criterion was applied, and, finally, whether statistical problems were described. Two independent raters categorized 83 articles based on these criteria. Results showed that a mixture of Fisher’s significance testing and Neyman-Pearson’s hypothesis testing was used in sport psychology. These approaches are often described as the ”hybrid model” or the ”null ritual.” Descriptions of test power were infrequent. The temporal analysis of articles revealed that effect sizes have been used increasingly in recent years. Ways to improve and standardize the use of statistical testing are presented and discussed.

Item Type:

Journal Article (Original Article)

Division/Institute:

07 Faculty of Human Sciences > Institute of Sport Science (ISPW)

UniBE Contributor:

Conzelmann, Achim

Subjects:

100 Philosophy > 150 Psychology
700 Arts > 790 Sports, games & entertainment

ISSN:

1612-5010

Publisher:

Hogrefe

Language:

German

Submitter:

Factscience Import

Date Deposited:

04 Oct 2013 15:16

Last Modified:

10 Aug 2015 11:08

Publisher DOI:

10.1026/1612-5010.16.2.43

Web of Science ID:

000265737700001

BORIS DOI:

10.7892/boris.33759

URI:

https://boris.unibe.ch/id/eprint/33759 (FactScience: 199379)

Actions (login required)

Edit item Edit item
Provide Feedback